\FloatBarrier
\begin{table}[H]\footnotesize \centering
\caption{Characteristics of treatment and control groups, 2001--2010 (I/II) }
\label{tab:treat-control_descript_incomers}
\begin{tabular}{lccc|ccc}
\toprule
  & \multicolumn{3}{c|}{\textbf{New in-coming taxpayers}}  & \multicolumn{3}{c}{\textbf{All taxpayers}}  \bigstrut[t]\\
  
  & {Treated} & {Control} & {Rest} & {Treated} & {Control} & {Rest} \\
      &       & \multicolumn{1}{l}{ 60-80\%} &       &       & \multicolumn{1}{l}{60-80\%} &  \\
\multicolumn{2}{l}{\textit{{Tax burden in \%}}} &       &       &       &       &  \\
\input{Table3-summarytable/summ_taxburden}


\multicolumn{3}{l}{\textit{Income (real, in 1000 CHF)}}   &     &       &       &  \\
\input{Table3-summarytable/summ_taxinc}


\multicolumn{3}{l}{\textit{Wealth (real, in 10,000 CHF)}}   &     &       &       &  \\
\input{Table3-summarytable/summ_taxwea}


\multicolumn{2}{l}{\textit{Employment status}} &       &       &       &       &  \\
\input{Table3-summarytable/summ_employment}

\multicolumn{2}{l}{\textit{Share of income from}} &       &       &       &       &  \\
\input{Table3-summarytable/summ_incshare} 
\vspace*{-0.2cm}  &  &       &       &       &       &  \\
\input{Table3-summarytable/summ_obs}
\bottomrule
\bottomrule
\end{tabular}
\justify
\textit{Note}: The table reports sample means for taxpayers who are in the treatment group, control group (with an income of 60-80\% of the regressive threshold), and remaining population of taxpayers. Means are reported separately for new incoming taxpayers (in which case the values refer to the year they first arrived in Obwalden) and the total stock of taxpayers, respectively. Standard deviations reported in parentheses.
NTR: net-of-average-tax rate, $(1-\bar{\tau})$.
Transfers: these include government transfers, such as pensions or unemployment benefits, as well as transfers from other households, namely alimony payments.
$^*$2006--2007; afterwards, cantonal income and wealth tax rates are identical for everyone, due to the 2008 flat rate tax reform.
\textit{Source}: Personal income and wealth tax data canton Obwalden, 2001--2010.
\end{table}



\begin{table}[H]\footnotesize \centering \ContinuedFloat
\caption{Characteristics of treatment and control groups, 2001--2010 (II/II) }
\begin{tabular}{lccc|ccc}
\toprule
  & \multicolumn{3}{c|}{\textbf{New in-coming taxpayers}}  & \multicolumn{3}{c}{\textbf{All taxpayers}}  \bigstrut[t]\\
  & {Treated} & {Control} & {Rest} & {Treated} & {Control} & {Rest} \\
      &       & \multicolumn{1}{l}{ 60-80\%} &       &       & \multicolumn{1}{l}{60-80\%} &  \\

\multicolumn{2}{l}{\textit{Household characteristics}} &       &       &       &       &  \\
\input{Table3-summarytable/summ_hhchar}
\vspace*{-0.2cm}  &  &       &       &       &       &  \\

\multicolumn{3}{l}{\textit{Tax savings from moving (real, in 1,000 CHF)}}  &   &       &       &  \\
\input{Table3-summarytable/summ_taxsavings}

\multicolumn{2}{l}{\textit{Driving distance to former residence }} &       &       &       &       &  \\
\input{Table3-summarytable/summ_dist}

\vspace*{-0.2cm}  &  &       &       &       &       &  \\
\input{Table3-summarytable/summ_obs}
\bottomrule
\bottomrule
\end{tabular}
\justify
\textit{Note}: The table reports sample means for taxpayers who are in the treatment group, control group (with an income of 60-80\% of the regressive threshold), and remaining population of taxpayers. Means are reported separately for new incoming taxpayers (in which case the values refer to the year they first arrived in Obwalden) and the total stock of taxpayers, respectively. Standard deviations reported in parentheses. \\
\textit{Source}: Personal income and wealth tax data canton Obwalden, 2001--2010.
\end{table}



\FloatBarrier